From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mga03.intel.com (mga03.intel.com [134.134.136.65]) by dpdk.org (Postfix) with ESMTP id 1B20B5960 for ; Tue, 21 Apr 2015 17:01:50 +0200 (CEST) Received: from fmsmga002.fm.intel.com ([10.253.24.26]) by orsmga103.jf.intel.com with ESMTP; 21 Apr 2015 08:01:31 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.11,616,1422950400"; d="scan'208";a="712956756" Received: from irsmsx152.ger.corp.intel.com ([163.33.192.66]) by fmsmga002.fm.intel.com with ESMTP; 21 Apr 2015 08:01:30 -0700 Received: from irsmsx105.ger.corp.intel.com ([169.254.7.178]) by IRSMSX152.ger.corp.intel.com ([169.254.6.7]) with mapi id 14.03.0224.002; Tue, 21 Apr 2015 16:01:29 +0100 From: "Ananyev, Konstantin" To: Olivier Matz , "dev@dpdk.org" Thread-Topic: [PATCH v5 08/12] mbuf: fix clone support when application uses private mbuf data Thread-Index: AQHQfBll8/9pXyv/+kW8vn2vSa6naJ1XjRzA Date: Tue, 21 Apr 2015 15:01:28 +0000 Message-ID: <2601191342CEEE43887BDE71AB9772582142095C@irsmsx105.ger.corp.intel.com> References: <1429544496-22532-1-git-send-email-olivier.matz@6wind.com> <1429610122-30943-1-git-send-email-olivier.matz@6wind.com> <1429610122-30943-9-git-send-email-olivier.matz@6wind.com> In-Reply-To: <1429610122-30943-9-git-send-email-olivier.matz@6wind.com> Accept-Language: en-IE, en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [163.33.239.180] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Subject: Re: [dpdk-dev] [PATCH v5 08/12] mbuf: fix clone support when application uses private mbuf data X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: patches and discussions about DPDK List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 21 Apr 2015 15:01:51 -0000 Hi Olivier, > -----Original Message----- > From: Olivier Matz [mailto:olivier.matz@6wind.com] > Sent: Tuesday, April 21, 2015 10:55 AM > To: dev@dpdk.org > Cc: Ananyev, Konstantin; zoltan.kiss@linaro.org; Richardson, Bruce; nhorm= an@tuxdriver.com; olivier.matz@6wind.com > Subject: [PATCH v5 08/12] mbuf: fix clone support when application uses p= rivate mbuf data >=20 > Add a new private_size field in mbuf structure that should > be initialized at mbuf pool creation. This field contains the > size of the application private data in mbufs. >=20 > Introduce new static inline functions rte_mbuf_from_indirect() > and rte_mbuf_to_baddr() to replace the existing macros, which > take the private size in account when attaching and detaching > mbufs. >=20 > Signed-off-by: Olivier Matz > Reviewed-by: Zoltan Kiss > --- > examples/vhost/main.c | 4 ++-- > lib/librte_mbuf/rte_mbuf.c | 2 +- > lib/librte_mbuf/rte_mbuf.h | 59 +++++++++++++++++++++++++++++++---------= ------ > 3 files changed, 43 insertions(+), 22 deletions(-) >=20 > diff --git a/examples/vhost/main.c b/examples/vhost/main.c > index 22d6a4b..195d82f 100644 > --- a/examples/vhost/main.c > +++ b/examples/vhost/main.c > @@ -138,7 +138,7 @@ > /* Number of descriptors per cacheline. */ > #define DESC_PER_CACHELINE (RTE_CACHE_LINE_SIZE / sizeof(struct vring_de= sc)) >=20 > -#define MBUF_EXT_MEM(mb) (RTE_MBUF_FROM_BADDR((mb)->buf_addr) !=3D (mb= )) > +#define MBUF_EXT_MEM(mb) (rte_mbuf_from_indirect(mb) !=3D (mb)) >=20 > /* mask of enabled ports */ > static uint32_t enabled_port_mask =3D 0; > @@ -1549,7 +1549,7 @@ attach_rxmbuf_zcp(struct virtio_net *dev) > static inline void pktmbuf_detach_zcp(struct rte_mbuf *m) > { > const struct rte_mempool *mp =3D m->pool; > - void *buf =3D RTE_MBUF_TO_BADDR(m); > + void *buf =3D rte_mbuf_to_baddr(m); > uint32_t buf_ofs; > uint32_t buf_len =3D mp->elt_size - sizeof(*m); > m->buf_physaddr =3D rte_mempool_virt2phy(mp, m) + sizeof(*m); > diff --git a/lib/librte_mbuf/rte_mbuf.c b/lib/librte_mbuf/rte_mbuf.c > index b013607..784ae8b 100644 > --- a/lib/librte_mbuf/rte_mbuf.c > +++ b/lib/librte_mbuf/rte_mbuf.c > @@ -129,7 +129,7 @@ rte_pktmbuf_init(struct rte_mempool *mp, >=20 > memset(m, 0, mp->elt_size); >=20 > - /* start of buffer is just after mbuf structure */ > + /* start of buffer is after mbuf structure and priv data */ > m->buf_addr =3D (char *)m + mbuf_size; > m->buf_physaddr =3D rte_mempool_virt2phy(mp, m) + mbuf_size; > m->buf_len =3D (uint16_t)buf_len; > diff --git a/lib/librte_mbuf/rte_mbuf.h b/lib/librte_mbuf/rte_mbuf.h > index 42db8e3..5c01c5b 100644 > --- a/lib/librte_mbuf/rte_mbuf.h > +++ b/lib/librte_mbuf/rte_mbuf.h > @@ -320,16 +320,40 @@ struct rte_mbuf { > }; > } __rte_cache_aligned; >=20 > +static inline uint16_t rte_pktmbuf_priv_size(struct rte_mempool *mp); > + > /** > - * Given the buf_addr returns the pointer to corresponding mbuf. > + * Return the mbuf owning the data buffer address of an indirect mbuf. > + * > + * @param mi > + * The pointer to the indirect mbuf. > + * @return > + * The address of the direct mbuf corresponding to buffer_addr. > */ > -#define RTE_MBUF_FROM_BADDR(ba) (((struct rte_mbuf *)(ba)) - 1) > +static inline struct rte_mbuf * > +rte_mbuf_from_indirect(struct rte_mbuf *mi) > +{ > + struct rte_mbuf *md; > + md =3D (struct rte_mbuf *)((char *)mi->buf_addr - > + sizeof(*mi) - rte_pktmbuf_priv_size(mi->pool)); > + return md; > +} Really puzzled, why do you remove prv_size from mbuf? Now if let say we attach mbuf (mi) with priv_size =3D=3D X to mbuf (md) wit= h priv_size =3D=3D Y, then rte_mbuf_from_indirect(mi) would return: mi->buf_addr - 128 - X, while= it should be: mi->buf_addr - 128 - Y. Another thing, rte_mbuf_from_indirect() and rte_mbuf_to_baddr() would proba= bly become a bit slower - as another level of indirection is added. Konstantin >=20 > /** > - * Given the pointer to mbuf returns an address where it's buf_addr > - * should point to. > + * Return the buffer address embedded in the given mbuf. > + * > + * @param md > + * The pointer to the mbuf. > + * @return > + * The address of the data buffer owned by the mbuf. > */ > -#define RTE_MBUF_TO_BADDR(mb) (((struct rte_mbuf *)(mb)) + 1) > +static inline char * > +rte_mbuf_to_baddr(struct rte_mbuf *md) > +{ > + char *buffer_addr; > + buffer_addr =3D (char *)md + sizeof(*md) + rte_pktmbuf_priv_size(= md->pool); > + return buffer_addr; > +} >=20 > /** > * Returns TRUE if given mbuf is indirect, or FALSE otherwise. > @@ -771,6 +795,7 @@ static inline struct rte_mbuf *rte_pktmbuf_alloc(stru= ct rte_mempool *mp) >=20 > /** > * Attach packet mbuf to another packet mbuf. > + * > * After attachment we refer the mbuf we attached as 'indirect', > * while mbuf we attached to as 'direct'. > * Right now, not supported: > @@ -784,7 +809,6 @@ static inline struct rte_mbuf *rte_pktmbuf_alloc(stru= ct rte_mempool *mp) > * @param md > * The direct packet mbuf. > */ > - > static inline void rte_pktmbuf_attach(struct rte_mbuf *mi, struct rte_mb= uf *md) > { > RTE_MBUF_ASSERT(RTE_MBUF_DIRECT(md) && > @@ -815,7 +839,8 @@ static inline void rte_pktmbuf_attach(struct rte_mbuf= *mi, struct rte_mbuf *md) > } >=20 > /** > - * Detach an indirect packet mbuf - > + * Detach an indirect packet mbuf. > + * > * - restore original mbuf address and length values. > * - reset pktmbuf data and data_len to their default values. > * All other fields of the given packet mbuf will be left intact. > @@ -823,22 +848,18 @@ static inline void rte_pktmbuf_attach(struct rte_mb= uf *mi, struct rte_mbuf *md) > * @param m > * The indirect attached packet mbuf. > */ > - > static inline void rte_pktmbuf_detach(struct rte_mbuf *m) > { > - const struct rte_mempool *mp =3D m->pool; > - void *buf =3D RTE_MBUF_TO_BADDR(m); > - uint32_t buf_len =3D mp->elt_size - sizeof(*m); > - m->buf_physaddr =3D rte_mempool_virt2phy(mp, m) + sizeof (*m); > + struct rte_mempool *mp =3D m->pool; > + uint32_t mbuf_size, buf_len; >=20 > - m->buf_addr =3D buf; > + mbuf_size =3D sizeof(struct rte_mbuf) + rte_pktmbuf_priv_size(mp); > + buf_len =3D rte_pktmbuf_data_room_size(mp); > + m->buf_addr =3D rte_mbuf_to_baddr(m); > + m->buf_physaddr =3D rte_mempool_virt2phy(mp, m) + mbuf_size; > m->buf_len =3D (uint16_t)buf_len; > - > - m->data_off =3D (RTE_PKTMBUF_HEADROOM <=3D m->buf_len) ? > - RTE_PKTMBUF_HEADROOM : m->buf_len; > - > + m->data_off =3D RTE_MIN(RTE_PKTMBUF_HEADROOM, (uint16_t)m->buf_len); > m->data_len =3D 0; > - > m->ol_flags =3D 0; > } >=20 > @@ -867,7 +888,7 @@ __rte_pktmbuf_prefree_seg(struct rte_mbuf *m) > * - free attached mbuf segment > */ > if (RTE_MBUF_INDIRECT(m)) { > - struct rte_mbuf *md =3D RTE_MBUF_FROM_BADDR(m->buf_addr); > + struct rte_mbuf *md =3D rte_mbuf_from_indirect(m); > rte_pktmbuf_detach(m); > if (rte_mbuf_refcnt_update(md, -1) =3D=3D 0) > __rte_mbuf_raw_free(md); > -- > 2.1.4